Modelling and COVID-19

Policy influence, lessons learned and pandemic preparedness

Sebastian Funk

Centre for Mathematical Modelling of Infectious Diseases

COVID-19 Modelling in the UK

Timeline of COVID-19 in the UK

Hospital admissions

  • LD1: Initial lockdown
  • LD2: Lockdown in response to autumn wave
  • LD3: Lockdown in response to emergence of alpha variant

Modelling COVID-19: early phase

Epidemiology and clinical picture

  • \(R_0\), serial interval, incubation period, length of stay, delay to death
  • asymptomatic proportion, case-fatality ratio, hospitalisation rate

Situational awareness

  • Growth rate
  • Short-term forecasts

Assessing the impact of early interventions

  • Travel restrictions
  • Contact tracing

Planning scenarios

  • Total number of hospitalisations and deaths
  • Impact of interventions

Modelling COVID-19: after initial phase

Situational awareness

  • Nowcasts (what is the latest picture)
  • Short-term forecasts

Planning scenarios

  • Relaxation of restrictions
  • Impact of new interventions (e.g. home testing)
  • Impact of vaccines

Epidemiological changes

  • Impact of new variants

Structure of policy advice in the UK

Modelling in the UK was underpinned by data

  • Large amounts of data in the UK, from the national to very local scale. Data publicly available, machine-readable, downloadable and ready for analysis.
  • Data (combined with models) was used to generate insights for policy.
  • Modelling can help inform data collection

Lessons learned and pandemic preparedness

Lessons learned: Data

Need infrastructure for rapid collection, cleaning, harmonisation, storage, sharing and publication of data.

Also a clear protocol for maximising the scope and amount of data that can be shared privately and publicly.

Lessons learned: Modelling

Policy makers should not have to rely on outputs from a single model.

Need to have sustainable structure in place that ensures constructive dialogue between policy makers and modellers.

Pandemic preparedness: Tools

We are working with WHO, CDC etc. to develop software tools for analysis and modelling of outbreak data.

Pandemic preparedness: Training

We are developing training material and courses for modelling in outbreaks.

https://epiverse-trace.github.io/learn.html
https://nfidd.github.io/nfidd/

Thank you